Design and Implementation of a Robust Acoustic Recognition System for Waterbird Species using TMS320C6713 DSK

Design and Implementation of a Robust Acoustic Recognition System for Waterbird Species using TMS320C6713 DSK

Amira Boulmaiz (Laboratory of Automatic and Signals of Annaba (LASA), Electronics department, Badji Mokhtar University, Annaba, Algeria), Djemil Messadeg (Laboratory of Automatic and Signals of Annaba (LASA), Electronics department, Badji Mokhtar University, Annaba, Algeria), Noureddine Doghmane (Laboratory of Automatic and Signals of Annaba (LASA), Electronics department, Badji Mokhtar University, Annaba, Algeria) and Abdelmalik Taleb-Ahmed (Laboratory of Automatic and Signals of Annaba (LASA), Electronics department, Badji Mokhtar University, Annaba, Algeria)
Copyright: © 2017 |Pages: 21
DOI: 10.4018/IJACI.2017010105
OnDemand PDF Download:
$37.50

Abstract

In this paper, a new real-time approach for audio recognition of waterbird species in noisy environments, based on a Texas Instruments DSP, i.e. TMS320C6713 is proposed. For noise estimation in noisy water bird's sound, a tonal region detector (TRD) using a sigmoid function is introduced. This method offers flexibility since the slope and the mean of the sigmoid function can be adapted autonomously for a better trade-off between noise overvaluation and undervaluation. Then, the features Mel Frequency Cepstral Coefficients post processed by Spectral Subtraction (MFCC-SS) were extracted for classification using Support Vector Machine classifier. A development of the Simulink analysis models of classic MFCC and MFCC-SS is described. The audio recognition system is implemented in real time by loading the created models in DSP board, after being converted to target C code using Code Composer Studio. Experimental results demonstrate that the proposed TRD-MFCC-SS feature is highly effective and performs satisfactorily compared to conventional MFCC feature, especially in complex environment.
Article Preview

Introduction

Algeria is home to a great diversity of wetlands, which are important sites for the wintering and the breeding of Palearctic migratory waterbirds. Tonga Lake, of El Kala National Park (northeast of Algeria), is a very special place for tens of thousands of different waterbirds species to overwinter or to a temporary halt. With an area of 2700 ha, Tonga lake is a refuge for over thirty-six different waterbird species and a major nesting site for Africa endangered waterbirds, namely: White-headed Duck (Oxyura leucocephala) and Ferruginous Duck (Aythya nyroca) which are threatened species ranked as “Endangered” and “Near Threatened” in the International union for conservation of nature red list, respectively. Currently, Tonga Lake and Garaet Hadj Tahar (northeast of Algeria) host the highest numbers of this species in Algeria and North Africa (Chettibi, 2013; Aissaoui et al., 2011).

One of primary element of ecological environment is environmental sound, which contains a great amount of rich information. Birds living in complex ecological environments can produce several types of sounds, which also contain a wealth of information closely related to the human survival environments. By examining, analyzing and recognizing bird sounds, pertinent information of birds can be acquired.

Apart from counting of individuals of each species, categorization of bird species and learning their life habits, automatic audio recognition of bird species has great potential significance in evaluating and predicting the surrounding living, as well as countless benefits on the protection development of ecological environment, which positively influence the development of economy and industry.

Recently and due to numerous analogies that exist between human speech and birdsongs (Acevedo et al., 2009), various techniques designed for automatic speech recognition have been adapted to automatic bird sound recognition as in (Lee el al., 2008; Chu & Blumstein, 2011; Jančovič & Kokuer, 2011; Patti, 2013). In a majority of these studies, databases used in the experiments consist on pretreated and enhanced sounds of different bird species (Stowell & Plumbley, 2014; Ptacek et al. 2016). However, in the real case where the birds are in their natural habitat, recordings of their sounds are subject to numerous environmental noises (e.g. sound of wind, rain, thunder, etc…) thus distorting the individual’s recognition operation.

The ultimate objective of this work is to develop a method based on embedded systems to survey the population of endangered waterbirds in natural habitat. The identification of different waterbird species can be done from an automatic recognition of their sounds. These kind of applications necessitate high-speed DSK (DSP Starter Kit). For this reason, in this study TMS320C6713 digital signal processor kit (DSK) has been used. The floating point TMS320C6713 DSP is a signal-processing platform of Texas Instrument. It features sufficient processing power for this project. As, it has a low current consumption to build an embedded system for bird monitoring in natural environment based on this DSP. Indeed, unlike the Raspberry board, that is rather a small computer for general use, with an operating system based on the Linux system, the TMS320C6713 DSP is a microprocessor optimized to perform different digital signal processing applications in real-time as in (Kumar, Singh, & Bhattacharya, 2015; Li, Bai, & Gao, 2015; López-Robles, Rodríguez-Rivas, Peralta-Sánchez, & Carranza-Castillo, 2015).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 8: 4 Issues (2017)
Volume 7: 2 Issues (2016)
Volume 6: 2 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing